Intermediate risk of choledocholithiasis: are we on the right path?

Introduction:

The risk of choledocholithiasis should be assessed in every patient undergoing cholecystectomy to define the next step. The American Society for Gastrointestinal Endoscopy proposed a stratified predictor scale of choledocholithiasis. Therefore, we aimed to describe our experience managing patients with an intermediate risk of choledocholithiasis according to the American Society for Gastrointestinal Endoscopy guidelines and the actual presence of bile duct stones in magnetic resonance cholangiopancreatography.

Methods:

A retrospective observational study with a prospective database was conducted. The analysis included sociodemographic data, laboratory values, and imaging. Bivariate, multivariate, and receiver operating characteristic analysis were performed.

Results:

Three hundred twenty-seven patients had an intermediate risk for choledocholithiasis. Half the patients were at least 65 years old. 24.77% were diagnosed with choledocholithiasis. Bile duct dilation was documented in only 3.06% of cases. Diagnosis of choledocholithiasis is associated with an age odds ratio (OR): 1.87 (P 0.02), alkaline phosphatase OR: 2.44 (P 0.02), and bile duct dilation greater than 6 mm OR: 14.65 (P 0.00).

Conclusions:

High variability in the accuracy of imaging techniques results in a large number of patients classified as intermediate risk without choledocholithiasis in cholangioresonance. Therefore, enhancing the criteria to define intermediate risk for patients in order to optimize resources is of paramount importance.

Introduction

Highlights

• A variety of covariates mislead a timely diagnosis and appropriate surgical management of common bile duct stones.

• High variability of imaging techniques results in a large number of patients classified as the intermediate risk of choledocholithiasis without actually having this condition.

• An improvement of the criteria to determine the intermediate risk of choledocholithiasis must be assessed in order to optimize medical and surgical resources.

The gallbladder is a small pouch that sits under the liver and stores bile1. Bile aids in the fat digestion process, during which cholesterol can deposit and form thick crystals (sludge) or stones (gallstones), formation of stones in the gallbladder is called cholelithiasis1,2. Additionally, gallstones can get plunged into the gallbladder, the cystic duct, or the common bile duct1,2. Cholelithiasis is a common disease, occurring in 5–22% of people in western countries, among 8–20% combined with common bile duct stone (CBDS)3,4. Choledocholithiasis is the presence of stones in the bile duct, commonly associated with cholelithiasis, with an incidence of 5–18%5,6. However, it also presents in patients with a history of cholecystectomy, with an incidence of 10–18%7. Choledocholithiasis is a frequent cause of hospitalization which may lead to cholangitis and gallstone pancreatitis3.

In order to avoid pancreatitis and cholangitis, every patient with cholelithiasis planning to undergo cholecystectomy should have its risk of preoperative choledocholithiasis figured, thus defining the next step, which can be either surgical or endoscopic5. Therefore, the American Society for Gastrointestinal Endoscopy (ASGE) proposed a strategy to assign the risk of choledocholithiasis and manage patients with symptomatic cholelithiasis based on clinical predictors such as dilated bile duct on an ultrasonographic image, age greater than 55 years, abnormal biochemical test, bilirubin greater than 4 mg/dl, clinical ascending cholangitis, and common bile duct stone (CBDS) that define risk into low, intermediate, and high-risk patients8,9. According to the ASGE criteria, patients with cholelithiasis have been classified as having a high, intermediate, or low risk of concomitant choledocholithiasis9. The most controversial group is the group of patients classified as an intermediate risk due to a wide range of diagnostic and management pathways8. According to the flow chart proposed by ASGE, patients with a low risk of choledocholithiasis should undergo cholecystectomy, while patients with a high risk of choledocholithiasis should undergo endoscopic retrograde pancreatography as a diagnostic and therapeutic procedure10,11. ASGE score is widely used worldwide, and the 2019 update is focused on reducing the amount of unnecessary endoscopic retrograde cholangiopancreatography (ERCP) but not the unnecessary use of magnetic resonance cholangiopancreatography (MRCP)12. Intermediate risk involves a large proportion of patients, therefore impacting prompt management and increasing health costs13.

The intermediate-risk group for choledocholithiasis constitutes a therapeutic challenge due to the possible need for presurgical biliary imaging12,13. MRCP has an 85–92% sensitivity and 93–97% specificity14,15, but it requires the provision of a resonator, the time to perform it, increased costs, and the interpretation of an experienced radiologist11–13. It is worth clarifying that the gold standard for the diagnosis of choledocholithiasis is ERCP, but with nonnegligible risks (5–10% morbidity) thus, it should be limited to high-risk patients14. On the other hand, despite its high sensibility, the use of endoscopic ultrasound (EUS) is not widely available and is limited to some institutions9,15,16.

The purpose of this study is to describe our experience managing patients with an intermediate risk of choledocholithiasis according to the ASGE guidelines and the confirmation of bile duct stones with MRCP.

Methods

Following and ensuring fully compliant criteria with the 2021 STROCSS (strengthening the reporting of cohort, cross-sectional and case–control studies in surgery) guidelines17 and following Institutional Review Board approval, all patients over 18 years old who were diagnosed with medium risk of choledocholithiasis according to the 2019 ASGE guidelines between January and December 2019 were registered. We exclude patients with micro-choledocholithiasis (Fig. 1). The present study was performed in a single The present study was performed in a single institution; two experienced radiologists performed all cholangioresonance interpretations. A retrospective analysis of patients who complied with inclusion criteria was made for the study. Ethical compliance with the Helsinki Declaration, current legislation on research Res. 008430-1993 and Res. 2378-2008 (Colombia), and the International Committee of Medical Journal Editors (ICMJE) were ensured under our Ethics and Research Institutional Committee (IRB) approval.

Figure 1.

Patients inclusion process. ASGE, American Society for Gastrointestinal Endoscopy.

Data obtained included patients’ demographics, liver function tests, additional diseases, imaging studies obtained, and their results. Descriptive statistics of all study parameters, such as the Marshall score and Tokyo scale, were provided18,19. Data were analyzed using STATA 17 software. Continuous data were summarized by their mean, standard deviation, median, and interquartile range. Categorical data were summarized by their frequency and proportion. Bivariate analysis was performed. Qualitative variables were analyzed using χ ² statistics (Fisher’s exact test when appropriate). Quantitative variables were analyzed based on normality, with Spearman’s or Pearson’s association’s correlation coefficients accordingly. Bivariate analysis between qualitative and quantitative variables was performed using the Mann–Whitney test and a t test for independent samples. Binary multivariate analysis was performed with significant variables found in the bivariate. Receiver operating characteristic analysis was used to determine cutoff points for relevant liver tests, and a new multivariate analysis was performed with new cutoff points.

Results

From January to December 2019, a total of 342 patients with the biliary disease were classified as having an intermediate risk of choledocholithiasis. Fifteen patients were excluded as they were initially misclassified as intermediate risk, obtaining a total of 327 patients with intermediate risk of choledocholithiasis according to the ASGE 2019 guidelines12. The median age was 65 [interquartile range (IQR) 20] years old, 70.72% were female and 29.28% were male. Half the patients had a time of symptomatology onset to consultation of 2 days. Within the paraclinical tests, leukocytosis was an important parameter, with 50% of patients presenting more than 10 745/mm³ (IQR 5640) leukocytes. The median total bilirubin was 1.415 mg/dl (IQR 1.505), usually at the expense of direct bilirubin with a median of 0.755 mg/dl (IQR 1.105). Liver profile was slightly altered in our population, with median aspartate aminotransferase (AST), alanine transaminase (ALT), and alkaline phosphatase (ALP) of 121 (IQR 291), 127 (IQR 287.5), and 165 IU/l (IQR 156), respectively (Table 1).

Table 1.

Quantitative descriptive characteristics.

Variable	Observation	Median	IQR
Age	327	65	21
Duration of symptoms	324	2	4
Leukocytes	326	10 745	5 640
Total bilirubin	324	1.415	1.505
Direct bilirubin	324	0.755	1.105
Indirect bilirubin	325	0.47	0.47
AST	321	121	291
ALT	320	127	287.5
Alkaline phosphatase	315	165	156

ALT, alanine transaminase; AST, aspartate aminotransferase; IQR, interquartile range.

All patients underwent magnetic resonance cholangiography (Table 2), while the vast majority (97.5%) had a previous hepatobiliary ultrasound. Of the 100% of patients taken to MRI cholangiography, only 24.7% were diagnosed with choledocholithiasis (Table 2). 37.6% of the patients presented with cholecystitis, among which 52.3% had a severity according to the Tokyo scale of II. 25.6% of patients presented with pancreatitis at the time of diagnosis, the majority (47.6%) with severity and organic failure according to the Marshall scale of II (Table 2). A significant percentage of patients (48.01%) had previous intra-abdominal surgery, 10.7% were type 2 diabetic, and 2.44% suffered from kidney disease.

Table 2.

Qualitative descriptive characteristics by sex.

	Sex
	Female	Male	Total, n (%)
Type 2 diabetes mellitus	23	12	35 (10.7)
Renal disease	6	2	8 (2.44)
Previous surgical intervention	113	44	157 (48.01)
Murphy sign positive	62	23	85 (25.99)
Cholecystitis	80	43	123 (37.6)
Tokyo 1	15	7	22 (17.8)
Tokyo 2	45	19	64 (52.3)
Tokyo 3	9	6	15 (12.19)
Pancreatitis	62	22	84 (25.6)
Marshall 0	6	0	6 (7.1)
Marshall 1	20	8	28 (33.3)
Marshall 2	29	11	40 (47.6)
Marshall 3	4	1	5 (5.9)
Marshall 4	2	1	3 (3.5)
Ultrasound	228	91	319 (97.5)
Magnetic resonance cholangiopancreatography	231	96	327 (100)
Choledocholithiasis diagnosis	53	28	81(24.7)
Total	231	96	327

Ultrasound findings indicated the presence of stones in the gallbladder or common bile duct, as well as the presence or absence of cholecystitis and bile duct dilation. According to these imaging findings, 49.38% of the patients had an ultrasound imaging diagnosis of cholelithiasis, 29.32% had concomitant cholecystitis, and only 1.54% of the patients with bile duct dilation were reported in this initial study (Table 3). Among the patients taken to MRCP, a diagnosis of cholelithiasis was made in 32.42% of the patients. 25.08% had concomitant cholelithiasis and cholecystitis. Regarding the dilation of the bile duct, the percentage was 3.06% (Table 4).

Table 3.

Ultrasound findings.

	Frequency	Percent
No data	5	1.54
Cholelithiasis	160	49.38
Cholelithiasis+cholecystitis	95	29.32
Cholelithiasis+bile duct dilation	13	4.01
Biliary sludge or microlithiasis	15	4.63
Cholecystitis	3	0.93
Bile duct dilation	5	1.54
Normal	25	7.72
Technical difficulties	3	0.93
Total	324	100.00

Table 4.

Magnetic resonance cholangiopancreatography findings.

	Frequency	Percent
Cholelithiasis	106	32.42
Cholelithiasis+cholecystitis	82	25.08
Choledocholithiasis	10	3.06
Cholecystocholedocholithiasis	69	21.10
Cholecystitis	3	0.92
Bile duct dilation	10	3.06
Pancreatitis	18	5.50
Normal	18	5.50
Neoplasm	9	2.75
Cholelithiasis+pancreatitis	2	0.61
Total	327	100.00

Bivariate analysis was performed including the ASGE score variables for intermediate risk. ALP OR: 1.00 (P 0.023), bile duct dilation greater than 6 mm OR: 12.48 (P=0.000), age odds ratio (OR): 1.02 (P=0.01), and total bilirubin OR: 1.00 (P=0.04) showed association with a statistically significant value in the diagnosis of choledocholithiasis. Multivariate analysis with those variables was performed, and the results are summarized in Table 5.

Table 5.

Bivariate analysis.

Variable	Odds ratio (95% CI)	P
Age	1.02 (1.004–1.03)	0.01
Bile duct >6 mm	12.48 (3.97–39.18)	0.000
Alkaline phosphatase	1.00 (1.002–1.003)	0.023
TGO	1.00 (0.99–1.007)	0.156
TGP	1.00 (0.99–1.001)	0.07
Total bilirubin	1.00 (0.81–1.24)	0.04

TGO, glutamic-oxaloacetic transaminase; TGP, glutamic pyruvic transaminase.

A receiver operating characteristic curve including age and ALP, was performed, seeking to define an appropriate cutoff value with acceptable sensitivity and specificity. The age value of 65 years old (S: 64.20%, E: 51.63%, AUC 0.59) and ALP 225 mg/dl [sensitivity (S): 50.2%, specificity (E): 70.46%, area under the curve (AUC) 0.62], and total bilirubin 1.6 mg/dl (S: 50%, E: 55%, AUC 0.62) were defined as cutoff points. In order to prove the statistical relationship of these cutoff values in the diagnosis of choledocholithiasis, multivariate regression was performed (Table 6). Age OR: 1.87 (P 0.02), ALP OR: 2.44 (P 0.02), and bile duct dilation greater than 6 mm OR: 14.65 (P 0.00) showed a stronger association with the diagnosis; total bilirubin did not show a statistically significant relationship with choledocholithiasis (P 0.391).

Table 6.

Multivariate analysis.

Variable	Odds ratio (95% CI)	P
Initial analysis
Age	1.01 (1.00–1.03)	0.02
Bile duct >6 mm	14.47 (4.48–46.7)	0.00
Alkaline phosphatase	1.01 (1.03–1.09)	0.02
TGO	–	–
TGP	1.05 (1.00–1.05	0.2
Total bilirubin	1.07 (0.85–1.3)	0.547
Cutoff analysis
Age	1.87 (1.06–3.31)	0.029
Alkaline phosphatase	2.44 (1.4–4.26)	0.002
Bile duct >6 mm	14.65 (4.5–47.9)	0.000

TGO, glutamic-oxaloacetic transaminase; TGP, glutamic pyruvic transaminase.

Discussion

Criteria for risk stratification proposed by the ASGE are frequently used worldwide. Following ASGE guidelines in 2010, patients are routinely categorized according to the probability of choledocholithiasis in low risk (<10%), intermediate risk (10–50%), and high risk (>50%)9. Patients with intermediate-risk benefit from additional biliary imaging20,21, and options for this include MRCP, EUS, and intraoperative cholangiography9. In their updated guidelines of 2019, ASGE suggests either EUS or MRCP confirm choledocholithiasis in intermediate-risk patients12. The differences between both classifications change the risk in six of our patients.

Moreover, the wide range of imaging studies paired with their varying sensitivity and specificity has enriched the discussion regarding subsequent steps after intermediate-risk classification. Meeralam et al.22 published a meta-analysis comparing EUS and MRCP diagnostic accuracy in which, albeit cost-effectiveness and adverse events were not taken into consideration, MRCP provided good diagnostic accuracy with a sensitivity and specificity of 87 and 92%, respectively. By protocol and availability at our institution, patients with intermediate risk for choledocholithiasis are assessed by MRCP. From 327 MRCP performed in this study, up to 24.9% were positive for choledocholithiasis. Jerónimo et al.23 found a choledocholithiasis frequency of 26.6% in patients with intermediate risk, which is comparable with our results. It is worth noting that the seemingly low frequency of choledocholithiasis in this group of patients may be related to lower sensitivity (33–71%) of MRCP in the setting of small stones (<3 mm)24,25, which was present in up to 4.72% of our patients according to ultrasound findings. Yet, Badger et al. found that most of the patients that underwent MRCP were followed with a more invasive test in 82% of cases, increasing the conflicting findings in the literature23.

Liver chemistries are measured as indirect markers of hepatobiliary disease26; standard markers include AST, ALT, ALP, and bilirubin as moderate predictors of choledocholithiasis in ASGE guidelines9. Aminotransferases including AST and ALT are enzymes involved in the transfer of amino groups of aspartate and alanine to ketoglutaric acid and referred to as transaminases26; regarding these predictors for intermediate risk, there is inconsistency in literature findings, some studies have shown that gamma-glutamyl transferase has the highest sensitivity, others demonstrate superiority in ALT for choledocholithiasis27. In our institution, median values of total bilirubin and ALP were within normal limits, whereas median values of transaminases were approximately three times over the cutoff level; gamma-glutamyl transferase levels were not measured. AST and ALT were elevated more than three-fold in at least 50% of patients with choledocholithiasis; nonetheless, no statistically significant correlation was found. Zare et al.28, in a prospective study performed on 350 patients for the assessment of liver function tests in the diagnosis of CBDSs in patients with cholecystitis, found ALT as an independent predictor of cholelithiasis (OR: 2; P=0.04). More recently, novel predictive scores are being proposed showing AUC of 0.77 and 0.76 for AST and ALT, respectively, in a cohort of 1089 patients29. Other descriptive studies have shown AST as the least sensitive parameter altered in ∼50.8% of the cases30. Thus, despite the discrepancy in liver enzyme specificity, the ongoing elevation of liver tests can provide support for deciding further imaging studies27.

Common bile duct (CBD) stones should be diagnosed on time, despite the lack of a consensus on the usefulness of liver enzymes measurement as in CBD stones; multiple studies suggest a positive association between ALP, ALT, and choledocholithiasis27. Isherwood et al.27 described the association between choledocholithiasis and echographic finding of common bile duct dilation more than 6 mm OR: 3.16 (P=0.06), and elevated values of ALP OR: 4.64 (P<0.00), and ALT (OR: 5.40, P<0.001). Nonetheless, despite our study, CBD dilation having an increased OR (12.48, IC: 3.97–39.18), transaminases did not show a statistically significant relationship as a diagnosis predictor.

Finally, our results and the comparison with the literature show the complexity of the diagnosis process, with important covariates that could mislead a timely diagnosis and could delay the surgical management of cholelithiasis; these expose the need for more specific and sensible predictive tools for CBD stones.

Among the limitations of this study are its retrospective nature, the lack of stronger associations between the variables and the outcome, and the scarcity of previous studies to compare our findings. Even though a considerable group of patients with cholelithiasis has concomitant intermediate choledocholithiasis, the risk scores available still give very loose recommendations regarding follow-up steps.

Conclusion

There is a high percentage of patients classified as an intermediate risk who do not have choledocholithiasis evidenced in cholangioresonance. There is persistently high variability in the accuracy of imaging techniques in intermediate-risk patients. CBD dilation, ALP, and age should be highly marked as predictive tools into the intermediate-risk score and could be a feasible tool to predict choledocholithiasis with acceptable sensitivity and specificity. Further studies are needed to confirm our results.

Ethics approval and consent to participate

All procedures performed in studies involving human participants were approved by the Institutional Review Board (Research Ethics Committee) and were in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. By institutional protocol, informed consent was reviewed for research purposes.

Sources of funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Patient consent

Written informed consent was obtained from the patient for the publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request.

Author contribution

F.G., L.M.R., D.C., C.E.R.C., and S.R.: conception of the idea; F.G., M.V., D.V., F.G., R.N., J.D.H., D.J., and R.E.N.-R.: data collection; F.G., M.V., D.V., F.G., R.N., J.D.H., D.J., R.E.N.-R., L.N., and S.R.: data analysis; F.G., L.M.R., D.C., C.E.R.C., R.E.N.-R., and S.R.: manuscript writing; F.G., L.M.R., D.C., C.E.R.C., M.V., D.V., F.G., R.E.N.-R., and S.R.: manuscript editing.

Conflicts of interest disclosure

The authors declare that they have no conflicts of interest.

Research registration unique identifying number (UIN)

None

Guarantor

Felipe Girón.

Data availability statement

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Provenance and peer review

Not commissioned, externally peer-reviewed.